Insulin delivery by implantable micropumps is one approach to the problem of restoration of normoglycemia sufficient to prevent the degenerative sequelae of diabetes. The objective of this proposal is the demonstration that insulin can be delivered by a diffusion based delivery system from an insulin suspension reservoir at both basal and augmented rates in pancreatectomized dogs without refilling the reservoir and thereby restore normoglycemia for longer periods of time than would be possible with a conventional solution reservoir. As proposed a 2 mL reservoir containing an insulin suspension (insulin crystals suspended in an aqueous buffer) could contain enough insulin for 1 - 2 years. With the support of NIH and others, many animal-months of experience with intraperitoneal insulin delivery (from solutions) have been obtained with the implanted controlled release micropump (CRM). A delivery program consisting of biphasic augmentation initiated approximately 15 min. before the start of a meal and basal delivery (4-7 U/day) at other times was found to be sufficient to sustain reasonable normoglycemia for over 7 months in the latest version. No aggregation has been observed with the use of Hoechst insulin (HOE 21P) which contains a surfactant to prevent aggregation. However, tissue reaction to the porous outlet does occur with time so that delivery performance deteriorates. Changing from a polyurethane to a silicone porous outlet has slowed the augmentation to compensate for the otherwise reduced insulin delivery. We propose to continue to investigate the performance of the CRM in diabetic dogs but now with the reservoir filled with a suspension rather than a solution. 24 hour glucose and insulin profiles will be used to assess in vivo delivery and determine the duration of reliable operation. The tissue after explanation will be examined histologically and the effect of the tissue reaction on in vivo delivery characteristics will be quantified in rats and dogs using 3H-inulin (not insulin) as a tracer as per Brown et al (52). Various pump outlet designs will be compared in their efficacy to control the tissue reaction. With such studies it will be possible to assess the potential benefits to the more than five million diabetics in North America to be derived from open loop insulin delivery systems.